What a deductible actually buys and sells
A deductible is the amount of each claim you retain. Raising it transfers a slice of risk from the insurer back to you, and the insurer returns part of the expected cost of that slice as a premium reduction. Only part, because the premium also carries expenses, commission, taxes and a risk margin, and the insurer keeps those on the portion it no longer pays.
That structural fact is why the deductible decision is not obvious. If the premium saving equalled the expected claims transferred, the choice would be a coin toss and you would take the higher deductible for the simplicity. It never does — it is either more than the expected transfer, in which case raising the deductible is straightforwardly good, or less, in which case you are buying back a small risk at a price.
The two questions worth asking are different, and people conflate them constantly.
Can I fund it? The break-even period answers this: how many years of the premium saving does it take to accumulate the extra amount you would have to produce at the scene of a claim. If the answer is two years and you have the cash today, the trade is comfortable. If it is nine years, the saving is not really funding anything.
Is it cheaper on average? The expected-cost comparison answers this: premium plus frequency times deductible, for each option. Whichever is smaller wins over a long run of similar years. That is the right test if you are indifferent to the variance and can absorb any single claim, which is the position of anyone with a comfortable emergency fund.
Two formulas, and the frequency where they meet
Break-even period. The extra you retain per claim is D₂ − D₁. The premium saving is P₁ − P₂ a year. Divide the first by the second and you have the number of claim-free years the saving needs to cover a single claim at the new deductible. That is all — no probability enters, which makes it a funding test rather than a value test.
Expected annual cost. Add the premium to the average deductible you expect to pay: E = P + λD, where λ is your expected number of deductible-bearing claims per year. This is the actuarially honest comparison, and it is linear in λ, which is why the chart above shows two straight lines.
Where they cross. Set the two expected costs equal: P₁ + λD₁ = P₂ + λD₂. Rearranging gives λ* = (P₁ − P₂)/(D₂ − D₁) — the premium saving divided by the extra deductible. Below that frequency the higher deductible is cheaper; above it the lower deductible is. Notice what λ* is: the reciprocal of the break-even period. A break-even of 2.94 years corresponds to a crossing frequency of 1 ÷ 2.94 = 0.34 claims a year, or one claim every 2.94 years. The two tests are the same statement read in different units, which is worth knowing so you do not treat them as independent evidence.
The practical reading is this: if you expect to claim less often than once every break-even period, the higher deductible is cheaper on average. That threshold is usually easy to beat for physical damage coverage, because at-fault collision and comprehensive claims are not annual events for most drivers — but you have to supply your own frequency, because it depends on your driving, your parking, your climate and your willingness to claim for small damage at all.
Worked example: $500 against $1,000
You are quoted $1,480 a year with a $500 deductible and $1,310 with a $1,000 deductible, on the same policy on the same day. You expect 0.07 deductible-bearing claims a year — about one every fourteen years — and you will hold the policy five years.
- Annual saving. 1,480 − 1,310 = $170.
- Extra exposure per claim. 1,000 − 500 = $500.
- Break-even period. 500 ÷ 170 = 2.94 years. Three claim-free years and the saving has funded the extra amount.
- Expected cost, lower deductible. 1,480 + 0.07 × 500 = 1,480 + 35 = $1,515.
- Expected cost, higher deductible. 1,310 + 0.07 × 1,000 = 1,310 + 70 = $1,380.
- Expected saving. 1,515 − 1,380 = $135 a year, and over five years 135 × 5 = $675.
- Crossing frequency. 170 ÷ 500 = 0.34 claims a year. You would need to claim roughly once every three years for the lower deductible to be the better buy.
On these numbers the higher deductible wins comfortably, and it wins by a wide margin: your assumed frequency of 0.07 is well below the 0.34 crossing point. Now test the assumption that matters. At 0.34 claims a year the two options cost the same. At 0.50 the lower deductible costs 1,480 + 0.50 × 500 = $1,730 and the higher costs 1,310 + 0.50 × 1,000 = $1,810, so the higher deductible is $80 worse. The whole decision lives in that one input, which is why the table sweeps it rather than trusting a single value.
Break-even years by premium saving and deductible step
| Annual premium saving | $250 step | $500 step | $750 step | $1,500 step |
|---|---|---|---|---|
| $50 | 5.00 | 10.00 | 15.00 | 30.00 |
| $75 | 3.33 | 6.67 | 10.00 | 20.00 |
| $100 | 2.50 | 5.00 | 7.50 | 15.00 |
| $150 | 1.67 | 3.33 | 5.00 | 10.00 |
| $200 | 1.25 | 2.50 | 3.75 | 7.50 |
| $300 | 0.83 | 1.67 | 2.50 | 5.00 |
Invert any cell to get the crossing claim frequency: a break-even of 3.33 years corresponds to 1 ÷ 3.33 = 0.30 claims a year, above which the lower deductible has the lower expected cost.
How to read the result honestly
Get both quotes from the same carrier on the same day. The comparison is only valid if nothing else differs. A saving that comes partly from a change of carrier, a change of limits or a lapsed discount is not a deductible saving, and the arithmetic here will mislead you.
Be honest about frequency, and remember it counts only deductible-bearing claims. A liability-only claim carries no deductible. A comprehensive claim for glass may carry a separate, often zero, deductible in some states and policies. What you are counting is collision and comprehensive claims you would actually file.
Small claims you would not file do not count. Many drivers with a $1,000 deductible would pay for $1,400 of damage themselves rather than file, because a claim can affect renewal pricing and a claim-free discount. If that describes you, the effective frequency for the higher deductible is lower than for the lower one — and the higher deductible is better than the calculation shows.
The variance matters even when the average does not. Expected cost is an average across many years. In any single year you either claim or you do not, and the higher deductible only pays off if you can produce the money without borrowing. If the extra amount would go on a credit card, add the interest to the cost of the option before deciding.
Check the interaction with a loan. The deductible comes straight off a total-loss settlement, which enlarges any shortfall against a loan balance. Run the total loss settlement calculator and the gap insurance need calculator at both deductibles: raising the deductible by $500 raises the gap by exactly $500 at every month of the schedule.
Consider dropping the coverage rather than raising the deductible. On an older vehicle, physical damage coverage can cost a meaningful fraction of what the car would settle for. If the actual cash value is close to the deductible, the coverage has little left to pay and the honest comparison is against no coverage at all, not against a larger deductible.
Do this on every deductible on the policy
Collision and comprehensive usually carry separate deductibles and separate premiums, and the trade is often better on one than on the other. Comprehensive claims — glass, theft, hail, animal strikes — have a different frequency profile from collision claims, and the premium saving per dollar of deductible differs too. Run the comparison twice with the coverage-level premiums from your declarations page rather than once with the policy total.
What this comparison leaves out
- The variance, entirely. Expected cost is a long-run average. It says nothing about the year you have two claims, which is exactly when the higher deductible hurts.
- Claim-severity truncation. A claim smaller than the higher deductible pays nothing under Option B and something under Option A. Modelling that properly needs a severity distribution, not a frequency; treat the frequency input as counting only claims larger than the higher deductible.
- Rating consequences of filing. A filed claim can affect renewal pricing and claim-free discounts, which makes the true cost of a claim higher than the deductible alone under either option.
- Waivers and vanishing deductibles. Some policies waive the deductible for glass repair, for a not-at-fault claim, or after a period of claim-free renewals. Those change the effective deductible without changing the number on the declarations page.
- Investment return on the saving. The break-even period assumes the saving sits idle. If it earns, the period shortens slightly; over a handful of years the effect is small relative to the uncertainty in the frequency.
- Premium drift. Both premiums change at renewal, and not necessarily by the same percentage. Re-run the comparison when they do.
Where this fits in the wider decision
The deductible is the smallest of the three levers on an auto premium, and it is worth putting it in proportion. Liability limits, coverage selection and driver assignment all move the number more, and the first of those should be decided on exposure rather than on price — the required limit comes out of your balance sheet in the auto liability coverage limit calculator, and it is not a budget line to be trimmed.
Where the deductible does matter disproportionately is on a policy that has been surcharged. A multiplier applies to the whole premium, so reducing the base by raising a deductible reduces the surcharge with it — the arithmetic is in the SR-22 cost calculator and the teen driver calculator, both of which multiply a base you can lower.
The same expected-cost machinery generalises well beyond auto policies. Homeowners deductibles work identically, except that wind and hail perils frequently carry a percentage deductible rather than a flat one, which changes the arithmetic completely — the retained amount scales with the dwelling limit rather than sitting at a fixed figure. That case is worked separately in the wind, hail and hurricane deductible calculator, and it is worth reading before assuming a homeowners deductible behaves like an auto one.
One closing point on how to think about it. The reason a higher deductible usually wins on expected cost is not that insurers are generous with the saving; it is that small claims are expensive to administer, so the insurer's cost of covering the first $500 of every claim is considerably more than the claims themselves. You are declining to buy the most expensive layer of coverage on the page. That is a good reason to take the higher deductible — provided you can actually pay it.